Eric's Value Averaging Page

Eric's Value Averaging Page

It is well known that dollar cost averaging has certain very desirable effects. First what is it? Dollar Cost Averaging refers to investing a fixed amount in some investment every period. The first benefit of this strategy is its simplicity. The second benefit is that it naturally helps you to take advantage of "timing the market." In other words, if you invest a fixed number of dollars every period, when the market is up, you buy fewer shares (when they may be overvalued). When the market is down, you buy more shares (when they may be undervalued). The cool thing about this is that by doing this, the average cost of the shares that you buy turns out to be less than the average price of the shares themselves! It's like magic, but it is simple and it works!

I read about an improvement from a professor at Harvard Business School (Dr. Michael E. Edleson) that he called "Value Averaging." (see "Value Averaging: A New Approach to Accumulation" in the AAII Journal, X, No. 7 (August 1988) and/or "Value Averaging The Safe and Easy Strategy for Higher Investment Returns", International Publishing Company, 1993). Instead of committing to invest a certain dollar amount every period, you would commit to grow your investment by that same dollar amount every period.

Example: Say you start with $1000 in your stock mutual fund and you commit to increase it by $1000 per month.

In a month, you check and find that it is then worth $1200. You have committed to increase it to $1000 + $1000 = $2000. Therefore, you invest an additional $800 to bring the total up to $2000.

The following month, you check and find that it is then worth $1500 (it has gone way down). You have committed to increase it to $2000 + $1000 = $3000. Therefore, you invest an additional $1500 to bring the total up to $3000.

The following month, you check and find that it is then worth the same $3000. You have committed to increase it to $3000 + $1000 = $4000. Therefore, you invest an additional $1000 to bring the total up to $4000.

The benefit of this methodology is that it amplifies the beneficial effects of dollar cost averaging. When the market is up, you buy even fewer shares. When the market is down, you buy even more shares.

A friend, Mike Barnes, pointed out a subtle flaw in this "Value Averaging" investment methodology. Since then, I have devised an elegant solution that significantly improves the methodology.

Mike correctly pointed out that the naturally expected appreciation of an investment would reduce (and eventually eliminate entirely) the contributions you would make under that scheme described above.

Here's the fix: Since you expect the investment to appreciate, you correct for it. Instead of committing to regularly grow your investment by a fixed amount (e.g., $1000) each period (e.g., each month), you commit to grow your investment by a certain percentage plus that fixed amount each period (e.g., by 1% plus $1000 each month). The key here is to choose an appropriate percentage. For Stocks and Stock Mutual Funds, since long term returns average about 12% annually, it seems natural to use that value, which very conveniently translates to a simple to calculate 1% per month (or 3% per quarter, etc.). If you wanted to get fancier, you might choose a percentage that is based on, for example, the rolling average return of that investment over the past year or some other period, but I suggest something simpler, like just settling on 1% a month.

Example: Say you start with $1000 in your stock mutual fund and you commit to increase it by 1% plus $1000 per month.

In a month, you check and find that it is then worth $1200. You have committed to increase it to ($1000 * 1.01) + $1000 = $2010. Therefore, you invest an additional $810 to bring the total up to $2010.

The following month, you check and find that it is then worth $1500 (it has gone way down). You have committed to increase it to ($2010 * 1.01) + $1000 = $3030. Therefore, you invest an additional $1530 to bring the total up to $3030.

The following month, you check and find that it is then worth the same $3030. You have committed to increase it to ($3030 * 1.01) + $1000 = $4060. Therefore, you invest an additional $1030 to bring the total up to $4060.

It turns out that in the steady state (i.e., the investment appreciating steadily at the specified rate (approx. 12% per annum in my example)), this methodology is exactly the same as Dollar Cost Averaging. If it varies either side of that rate, this methodology exaggerates the desirable results of Dollar Cost Averaging, making this a superior systematic investment methodology (if you don't take into account the "cash drag" and "cost of credit" discussed below).

For a recent journal article on this subject, see the following:

Paul S. Marshall, "A Statistical Comparison of Value Averaging vs. Dollar Cost Averaging and Random Investment Techniques," Journal of Financial and Strategic Decisions, Spring 2000, pp. 87-99.

Unfortunately, the Marshall article suffers from the same problem that the original Edleson work did: in comparing dollar cost averaging with value averaging, it didn't take into account the "cash-drag" that value averaging suffers when it invests less in a period than dollar cost averaging does. It also doesn't take into account the cost of credit when value averaging invests more in a period than dollar cost averaging does. Unfortunately, when these two effects are taken into account, Value Averaging loses much of its luster (as it also loses much, or all of its supposed advantage over dollar cost averaging).

Eric's Financial Page
Eric's personal Home Page

To send Eric an E-Mail, click here.

This page last updated 09/01/03